Cloning and functional expression of a novel degenerin-like Na+ channel gene in mammals

Abstract

1. A degenerate polymerase chain reaction (PCR) homology screening procedure was applied to rat brain cDNA in order to identify novel genes belonging to the amiloride-sensitive Na+ channel and degenerin (NaC/DEG) family of ion channels. A single gene was identified that encodes a protein related to but clearly different from the already cloned members of the family (18-30 % amino acid sequence identity). Phylogenetic analysis linked this protein to the group of ligand-gated channels that includes the mammalian acid-sensing ion channels and the Phe-Met-Arg-Phe-amide (FMRFamide)-activated Na+ channel. 2. Expression of gain-of-function mutants after cRNA injection into Xenopus laevis oocytes or transient transfection of COS cells induced large constitutive currents. The activated channel was amiloride sensitive (IC50, 1.31 microM) and displayed a low conductance (9-10 pS) and a high selectivity for Na+ over K+ (ratio of the respective permeabilities, PNa+/PK+ >= 10), all of which are characteristic of NaC/DEG channel behaviour. 3. Northern blot and reverse transcriptase-polymerase chain reaction (RT-PCR) analysis revealed a predominant expression of its mRNA in the small intestine, the liver (including hepatocytes) and the brain. This channel has been called the brain-liver-intestine amiloride-sensitive Na+ channel (BLINaC). 4. Corresponding gain-of-function mutations in Caenorhabditis elegans degenerins are responsible for inherited neurodegeneration in the nematode. Besides the BLINaC physiological function that remains to be established, mutations in this novel mammalian degenerin-like channel might be of pathophysiological importance in inherited neurodegeneration and liver or intestinal pathologies.

Figure 3. Tissue distribution of BLINaC mRNA in different mouse and rat tissues

A, RT-PCR experiments performed with RNA isolated from different mouse tissues (indicated on top of each lane). PCR products were hybridized with an α-³²P-labelled mouse BLINaC cDNA probe (upper panel). The expected PCR product size is given on the left. Amplification of GAPDH was used for control (lower panel, ethidium bromide staining). B, analysis of BLINaC expression on different rat tissues (indicated on top) by RT-PCR. Specificity was confirmed by hybridization of the expected 365 bp PCR product with an α-³²P-labelled rat BLINaC cDNA probe (upper panel). A control amplification of β-actin was performed in parallel (lower panel, ethidium bromide staining). C, detection of BLINaC transcript on mouse brain, liver and small intestine poly A⁺ RNA (5 μg per lane) by Northern blot analysis. The probe used corresponds to an α-³²P-labelled mouse BLINaC cDNA fragment. The filter was reprobed with a GAPDH probe as an RNA loading control (lower panel). A transcript of 2.1 kb was strongly detected in liver while a smaller transcript of 1.6 kb was found in small intestine. D, BLINaC mRNA expression in mouse liver and in freshly prepared hepatocytes was assessed by RT-PCR experiments. Specific amplification in liver and pure hepatocytes of a 737 bp product (upper panel, Southern blot) and β-actin control amplification (lower panel, ethidium bromide staining) are shown. The control corresponds to a PCR without cDNA.

Figure 1. Sequence analysis of the BLINaC protein

A, sequence comparison between rat BLINaC and the closest relative ASIC1 proteins. The two putative transmembrane regions MI and MII are indicated by bold lines above the sequence. External cysteines are marked with filled circles. The mutated alanine involved in the gain-of-function just before the TMII is indicated by an asterisk. Identical and similar residues are printed white on black and black on grey, respectively. The sequences were aligned using the Genetics Computer Group (GCG) Pileup program (Madison, WI, USA), with minor manual corrections when necessary. B, Kyte and Doolittle hydropathicity plot. Hydropathy values (positive is hydrophobic, negative is hydrophilic) represent the average over a 19 residues window. Amino acid numbering and putative transmembrane domains are given below the plot. C, BLINaC putative membrane topology deduced from its hydropathy profile and from the data obtained on other members of the NaC/DEG family. The extracellular loop represents most of the protein while the intracellular C-terminal part is extremely short.

Figure 2. Sequence comparison between BLINaC and the other NaC/DEG family members

A, phylogenetic analysis. ASICs are acid-sensing ion channel subunits including ASIC1, ASIC2 (previously named MDEG1) and ASIC3 (previously named DRASIC). FaNaC is the FMRFamide-activated Na⁺ channel. MEC-4, MEC-10, DEG-1, DEL-1, UNC-8 and UNC-105 are Caenorhabditis elegans degenerins. α, β and γ ENaC are subunits of the epithelial Na⁺ channel and δNaC is an αENaC-like subunit. dGNaC1 (ripped pocket) and dmdNaC1 (pickpocket) are Drosophila members of the NaC/DEG family. The phylogenetic tree was established using the GCG Distances program with Kimura substitution followed by the GCG Growtree program with the UPGMA option, from an alignment obtained with the GCG Pileup program. Phylogenetic analysis clearly associates BLINaC with the subgroup of ligand-gated channels that comprise ASICs and FaNaC. B, percentage of amino acid sequence identity between BLINaC and other members of the amiloride-sensitive Na⁺ channel/degenerin family. Percentages were calculated using the GCG Distances program without correction for multiple substitutions. Accession numbers for ASIC1, ASIC3 (previously named DRASIC), FaNaC, MEC-4, rat αENaC, δNaC and dGNaC1 are U94403, AF013598, X92113, U53669, X70521, U38254 and Y16240, respectively. The overall identity with the other NaC/DEG channels remains low (below or equal to 30 %) while much higher identity may exist locally (e.g. in the second transmembrane domain region).

Figure 6. Localization of the BLINaC probe on WMP murine chromosomes

A, R-banded metaphase chromosomes show the fluorescent hybridization signals on chromosome 3 belonging to the (3;12) chromosomes. B, the corresponding localization (3E-3F1) is shown on the idiogram of the G-banded (3;12) Robertsonian translocation.

Figure 4. Properties of the rat wild-type (WT) and mutated BLINaC expressed in Xenopus oocytes

A, comparison of the amiloride-sensitive current (I_amiloride) recorded from oocytes injected with WT or A443 mutants (indicated below the columns) of BLINaC using the two-microelectrode voltage clamp. Currents were recorded at a holding potential of -70 mV in the presence or in the absence of 1 mM amiloride. CTR corresponds to water-injected oocytes. Columns and bars represent the means and s.e.m., the number of analysed oocytes is given above the columns. B, current-voltage relationship of the amiloride-sensitive current (1 mM amiloride) recorded from oocytes injected with WT BLINaC cRNA (3 ng oocyte⁻¹; typical recording) and A443T mutant cRNA (0.25 ng oocyte⁻¹; mean of 3 oocytes) obtained by voltage ramps from -150 to +50 mV in Na⁺-containing medium (ND96 solution). C, effect of Na⁺ replacement by Li⁺ or K⁺ in the external medium on the amiloride-sensitive current of the A443T mutant recorded at -70 mV. D, concentration-dependent inhibition of the A443T mutant currents by amiloride (•), benzamil (▴) and EIPA (▪) and inhibition of the WT BLINaC constitutive current by amiloride (_•). The currents were measured at -70 mV in the ND96 bathing solution. Each point represents the mean value from four oocytes.

Figure 5. Single-channel properties of BLINaC A443 gain-of-function mutants transiently expressed in COS cells

A, typical traces of single-channel activity in outside-out patches recorded at a membrane potential of -50 mV for four BLINaC gain-of-function mutants; c and o indicate the closed and opened states, respectively. B, effect of amiloride (1 and 10 μm) on the single-channel current recorded at -50 mV from an outside-out patch containing two A443T mutant channels. The amiloride block is reversible and concentration dependent. C, mean single-channel current-voltage relationships of the four BLINaC gain-of-function mutants recorded from outside-out patches in Na⁺-containing medium (140 mM Na⁺ in the medium). The estimated reversal potential was at least larger than 58 mV (P_Na/P_K≥ 10). Data were collected from two (A443T, □), three (A443S, •; and A443F, ▪) and four (A443C; ○) different patches. Unitary conductance corresponding to each mutant was estimated from the slope between -50 and 0 mV. D, single-channel current traces at different membrane potentials (indicated to the left) obtained from an outside-out patch containing two A443T mutant channels in the presence of 140 mM Na⁺ in the bathing solution. Amiloride application (1 μm) can block the channel at all potentials.